This proposal describes studies of the zeste-white-10 (zw-10) locus of Drosophila. Genetic data indicate that mutant alleles of this locus cause nondisjunction and breakage of mitotic chromosomes. Metaphase nuclei in larval ganglial cells are often hyperploid and sister chromatids are often prematurely separated. This centromere behavior is the likely precursor to nondisjunction and breakage. Some of the proposed experiments will continue the genetic and cytological characterization of mutant alleles of this locus. Experiments with temperature sensitive alleles will be done to ask whether the function is required continuously throughout the cell cycle. Other experiments will ask whether the nuclear divisions of the embryo are maternally programmed while divisions within imaginal discus and histoblasts are dependent upon the zygotic nucleus. The relationship of chromosome breakage to nondisjunction will be studied by asking 1) whether breakage, like nondisjunction, is a centromere related event and 2) whether nondisjunction and breakage are alternative outcomes of the same preconditions. Breakage will also be studied cytologically. These studies will employ chromosomes carrying cytological landmarks to assist in identifying sites of breakage. Additional cytological studies will be done using indirect immunofluorescence and differential interference contrast microscopy of spindle structure in zw-10 nuclei. The second major focus of the proposed research is a molecular analysis of zw-10. One objective of molecular studies is to test the hypothesis that the zw-10 product is a protein which acts to preserve sister centromere cohesion until anaphase separation. The zw-10+ DNA will be isolated and spliced to a prokaryotic promoter. This recombinant DNA will be introduced into a bacterial host in order to produce a fusion protein. Antibodies to the protein will be fluorescence coupled and used in turn to locate the zw-10+ product in situ. During the initial studies of zw-10, evidence was found for a related gene function nearby; the presence of an extra copy of this second gene partially compensates for a mutant zw-10 genotype. This observation ultimately led to the recognition that the zw interval is an ancestral duplication in the form of a reverse repeat. Genetic and molecular experiments will be done to test this notion and identify the zw-10 cognate. The cognate function will then be studied in the way described here for zw-10.